Ultrasonic Echo Object Recognition in Low-Light and Obscured Scenes

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing object recognition technologies based on image recognition are prone to errors due to variations in scenarios, affecting accuracy and security of terminals.

Innovation Solution

An object recognition method using ultrasonic echoes, where ultrasonic signals are transmitted and reflected to extract features, which are then converted to a target dimension and translated into image information, ensuring accuracy and security across various scenarios.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If visible light or electromagnetic waves are used for detection, then detection speed is improved, but detection accuracy deteriorates in environments with smoke, dust, or darkness

Engineering Contradiction:
Improvedetection speedVSAvoiddetection accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent changes the detection parameter from visible light to ultrasonic waves, which have different physical properties that allow them to penetrate smoke, dust, and darkness effectively. This parameter change resolves the contradiction by maintaining detection accuracy in adverse environments while preserving the speed advantage of electromagnetic detection methods.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If ultrasonic waves are used for detection, then detection accuracy in adverse environments is improved, but detection speed deteriorates compared to visible light

Engineering Contradiction:
Improvedetection accuracyVSAvoiddetection speed
Core Design Contradiction:
Measurement precisionVSSpeed

Solution Approach 1:

The patent employs periodic ultrasonic wave emission and reception cycles to achieve rapid successive detections. By using periodic action with optimized pulse intervals, the system maintains high detection speed while utilizing ultrasonic waves for accurate detection in smoke, dust, and darkness, thus resolving the speed disadvantage.

Inventive Principle:
Principle #19Periodic action

3Device complexity

If traditional object recognition methods are used, then system complexity is reduced, but adaptability to different environments and objects deteriorates

Engineering Contradiction:
Improvesystem complexityVSAvoidenvironmental adaptability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent creates a universal detection system using ultrasonic waves that can detect various objects (people, vehicles, obstacles) in different environments (smoke, dust, darkness, open spaces). The single ultrasonic detection apparatus performs multiple detection functions, achieving high adaptability without proportionally increasing system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The method ensures accurate object recognition by translating physical information from ultrasonic echoes into image information, reducing sensitivity to scenario changes and enhancing terminal security.

Implementation Method 1

an object recognition method which recognizes an object by using ultrasonic echo

Methodology Applied
Scientific EffectUltrasonic echo: Echo

Implementation Method 2

emitting ultrasonic waves toward a detection target using a transmission ultrasonic sensor

Methodology Applied
Scientific EffectUltrasonic wave reflection: Reflection

Data Source

PatentEP4198559B1Object recognition method and apparatus based on ultrasonic echo, and storage medium
Publication Date: 2026.05.13 TENCENT TECHNOLOGY (SHENZHEN) CO LTD
  • EP4198559B1 patent drawingFigure 1
  • EP4198559B1 patent drawingFigure 2
  • EP4198559B1 patent drawingFigure 3

AI summary

An object recognition method and apparatus based on an ultrasonic echo, and a storage medium, which are applied to the computer vision technology of artificial intelligence. Said method comprises: transmitting, by means of an acoustic wave transmission apparatus of a terminal, an ultrasonic signal to an object to be recognized (301); then receiving, by means of an acoustic wave reception apparatus, an echo signal reflected by said object (302); extracting an ultrasonic echo feature of the echo signal (303); performing feature dimension conversion on the ultrasonic echo feature to obtain a target dimension feature (304); and then performing image translation processing to obtain object image information corresponding to said object (305). With respect to image acquisition, sufficient ambient light is required to achieve accurate imaging, and transmission of an ultrasonic signal and reception of a reflected echo signal are less sensitive to changes in a scene; therefore, in a diversity scene where image acquisition is applicable or not applicable, object image information of an object to be recognized can be accurately translated by means of physical information carried by the ultrasonic echo, ensuring the accuracy of object recognition and improving the security of a terminal.